unicode.rb in characteristics-0.7.0

- old
+ new

@@ -89,10 +89,80 @@
     0x2067,
     0x2068,
     0x2069,
   ].freeze
 
+  VARIATION_SELECTORS = [
+    *0x180B..0x180D,
+    *0xFE00..0xFE0F,
+    *0xE0100..0xE01EF,
+  ].freeze
+
+  TAGS = [
+    0xE0001,
+    *0xE0020..0xE007F,
+  ].freeze
+
+  NONCHARACTERS = [
+    *0xFDD0..0xFDEF,
+    0xFFFE,   0xFFFF,
+    0x1FFFE,  0x1FFFF,
+    0x2FFFE,  0x2FFFF,
+    0x3FFFE,  0x3FFFF,
+    0x4FFFE,  0x4FFFF,
+    0x5FFFE,  0x5FFFF,
+    0x6FFFE,  0x6FFFF,
+    0x7FFFE,  0x7FFFF,
+    0x8FFFE,  0x8FFFF,
+    0x9FFFE,  0x9FFFF,
+    0xAFFFE,  0xAFFFF,
+    0xBFFFE,  0xBFFFF,
+    0xCFFFE,  0xCFFFF,
+    0xDFFFE,  0xDFFFF,
+    0xEFFFE,  0xEFFFF,
+    0xFFFFE,  0xFFFFF,
+    0x10FFFE, 0x10FFFF,
+  ].freeze
+
+  IGNORABLE = [
+    0x00AD,
+    0x034F,
+    0x061C,
+    *0x115F..0x1160,
+    *0x17B4..0x17B5,
+    *0x180B..0x180E,
+    *0x200B..0x200F,
+    *0x202A..0x202E,
+    *0x2060..0x206F,
+    0x3164,
+    *0xFE00..0xFE0F,
+    0xFEFF,
+    0xFFA0,
+    *0xFFF0..0xFFF8,
+    *0x1BCA0..0x1BCA3,
+    *0x1D173..0x1D17A,
+    *0xE0000..0xE0FFF,
+  ].freeze
+
+  KDDI = [
+    *0xE468..0xE5DF,
+    *0xEA80..0xEB8E,
+  ].freeze
+
+  SOFTBANK = [
+    *0xE001..0xE05A,
+    *0xE101..0xE15A,
+    *0xE201..0xE25A,
+    *0xE301..0xE34D,
+    *0xE401..0xE44C,
+    *0xE501..0xE53E,
+  ].freeze
+
+  DOCOMO = [
+    *0xE63E..0xE757,
+  ].freeze
+
   attr_reader :category
 
   def initialize(char)
     super
 
@@ -140,31 +210,45 @@
 
   def bidi_control?
     @is_valid && BIDI_CONTROL.include?(@ord)
   end
 
+  # unicode specific
+
+  def variation_selector?
+    @is_valid && VARIATION_SELECTORS.include?(@ord)
+  end
+
+  def tag?
+    @is_valid && TAGS.include?(@ord)
+  end
+
+  def noncharacter?
+    @is_valid && NONCHARACTERS.include?(@ord)
+  end
+
+  def ignorable?
+    @is_valid && IGNORABLE.include?(@ord)
+  end
+
+  # emoji
+
   def kddi?
     @is_valid &&
     encoding_has_kddi? &&
-    ( @ord >= 0xE468 && @ord <= 0xE5DF ||
-      @ord >= 0xEA80 && @ord <= 0xEB8E )
+    KDDI.include?(@ord)
   end
 
   def softbank?
     @is_valid &&
     encoding_has_softbank? &&
-    ( @ord >= 0xE001 && @ord <= 0xE05A ||
-      @ord >= 0xE101 && @ord <= 0xE15A ||
-      @ord >= 0xE201 && @ord <= 0xE25A ||
-      @ord >= 0xE301 && @ord <= 0xE34D ||
-      @ord >= 0xE401 && @ord <= 0xE44C ||
-      @ord >= 0xE501 && @ord <= 0xE53E )
+    SOFTBANK.include?(@ord)
   end
 
   def docomo?
     @is_valid &&
     encoding_has_docomo? &&
-    ( @ord >= 0xE63E && @ord <= 0xE757 )
+    DOCOMO.include?(@ord)
   end
 
   private
 
   def encoding_has_c1?